diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool3d_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool3d_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..a592ee91b159c80f4464061ef188c37ab18a83db --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_adaptive_avg_pool3d_backward_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _adaptive_avg_pool3d_backward_out(const at::Tensor & grad_output, const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor adaptive_avg_pool3d_backward_cpu(const at::Tensor & grad_output, const at::Tensor & self); +TORCH_API at::Tensor adaptive_avg_pool3d_backward_cuda(const at::Tensor & grad_output, const at::Tensor & self); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_autocast_to_reduced_precision_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_autocast_to_reduced_precision_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..bb14e5b7deb4df6cc93d24772e3cb20579b2e808 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_autocast_to_reduced_precision_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor _autocast_to_reduced_precision(const at::Tensor & self, bool cuda_enabled, bool cpu_enabled, at::ScalarType cuda_dtype, at::ScalarType cpu_dtype); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_chunk_cat_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_chunk_cat_native.h new file mode 100644 index 0000000000000000000000000000000000000000..e42d4984623b1a283051df1f1df1e7f8dd531b76 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_chunk_cat_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _chunk_cat(at::TensorList tensors, int64_t dim, int64_t num_chunks); +TORCH_API at::Tensor & _chunk_cat_out(at::TensorList tensors, int64_t dim, int64_t num_chunks, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_conv_depthwise2d_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_conv_depthwise2d_native.h new file mode 100644 index 0000000000000000000000000000000000000000..88a15f35ead916c1d1ce5e35a80d42fb95cc55f8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_conv_depthwise2d_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor conv_depthwise2d_cuda(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation); +TORCH_API const at::Tensor & conv_depthwise2d_cuda_out(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, const at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cslt_sparse_mm_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cslt_sparse_mm_native.h new file mode 100644 index 0000000000000000000000000000000000000000..82976bda45d2fb66f9e93e300157db2731009c9b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_cslt_sparse_mm_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _cslt_sparse_mm(const at::Tensor & compressed_A, const at::Tensor & dense_B, const c10::optional & bias={}, const c10::optional & alpha={}, c10::optional out_dtype=c10::nullopt, bool transpose_result=false, int64_t alg_id=0); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_euclidean_dist_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_euclidean_dist_native.h new file mode 100644 index 0000000000000000000000000000000000000000..494c9aa65b15d61addc64c914ec419a4d826c9fc --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_euclidean_dist_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _euclidean_dist(const at::Tensor & x1, const at::Tensor & x2); +TORCH_API at::Tensor & _euclidean_dist_out(const at::Tensor & x1, const at::Tensor & x2, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_floor_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_floor_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d79a0b653d06835297e4a1cf1f2a12c448b893c1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_floor_cpu_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::vector _foreach_floor(at::TensorList self); +TORCH_API void _foreach_floor_(at::TensorList self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid.h new file mode 100644 index 0000000000000000000000000000000000000000..00e2c0117307327ee6dcc2e55d5126c6e1aaf69f --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_foreach_sigmoid.h @@ -0,0 +1,44 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_foreach_sigmoid(Tensor[] self) -> Tensor[] +inline ::std::vector _foreach_sigmoid(at::TensorList self) { + return at::_ops::_foreach_sigmoid::call(self); +} + +// aten::_foreach_sigmoid_(Tensor(a!)[] self) -> () +inline void _foreach_sigmoid_(at::TensorList self) { + return at::_ops::_foreach_sigmoid_::call(self); +} + +// aten::_foreach_sigmoid.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_sigmoid_out(at::TensorList out, at::TensorList self) { + return at::_ops::_foreach_sigmoid_out::call(self, out); +} +// aten::_foreach_sigmoid.out(Tensor[] self, *, Tensor(a!)[] out) -> () +inline void _foreach_sigmoid_outf(at::TensorList self, at::TensorList out) { + return at::_ops::_foreach_sigmoid_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_functional_sym_constrain_range_for_size_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_functional_sym_constrain_range_for_size_native.h new file mode 100644 index 0000000000000000000000000000000000000000..927633ade2492489ac7834af347c86e4a7d5b450 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_functional_sym_constrain_range_for_size_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor _functional_sym_constrain_range_for_size(const at::Scalar & size, c10::optional min, c10::optional max, const at::Tensor & dep_token); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_solve_ex_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_solve_ex_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..ab438ecdd24b4d856d84618445865732b38d57d5 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_linalg_solve_ex_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _linalg_solve_ex { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, bool, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_linalg_solve_ex") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_linalg_solve_ex(Tensor A, Tensor B, *, bool left=True, bool check_errors=False) -> (Tensor result, Tensor LU, Tensor pivots, Tensor info)") + static ::std::tuple call(const at::Tensor & A, const at::Tensor & B, bool left, bool check_errors); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, const at::Tensor & B, bool left, bool check_errors); +}; + +struct TORCH_API _linalg_solve_ex_result { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, bool, bool, at::Tensor &, at::Tensor &, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_linalg_solve_ex") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "result") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_linalg_solve_ex.result(Tensor A, Tensor B, *, bool left=True, bool check_errors=False, Tensor(a!) result, Tensor(b!) LU, Tensor(c!) pivots, Tensor(d!) info) -> (Tensor(a!) result, Tensor(b!) LU, Tensor(c!) pivots, Tensor(d!) info)") + static ::std::tuple call(const at::Tensor & A, const at::Tensor & B, bool left, bool check_errors, at::Tensor & result, at::Tensor & LU, at::Tensor & pivots, at::Tensor & info); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & A, const at::Tensor & B, bool left, bool check_errors, at::Tensor & result, at::Tensor & LU, at::Tensor & pivots, at::Tensor & info); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_native.h new file mode 100644 index 0000000000000000000000000000000000000000..497dfcda9d90ab77fa3d6e4583d6ca0e245ecaed --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_masked_scale_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _masked_scale_out(const at::Tensor & self, const at::Tensor & mask, double scale, at::Tensor & out); +TORCH_API at::Tensor masked_scale_cuda(const at::Tensor & self, const at::Tensor & mask, double scale); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_buffer_copy_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_buffer_copy_native.h new file mode 100644 index 0000000000000000000000000000000000000000..fe78e3a05dce7a97b4ece22777afc5c1212bda50 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_nested_view_from_buffer_copy_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _nested_view_from_buffer_copy_out(const at::Tensor & self, const at::Tensor & nested_size, const at::Tensor & nested_strides, const at::Tensor & offsets, at::Tensor & out); +TORCH_API at::Tensor _nested_view_from_buffer_copy(const at::Tensor & self, const at::Tensor & nested_size, const at::Tensor & nested_strides, const at::Tensor & offsets); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pad_packed_sequence.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pad_packed_sequence.h new file mode 100644 index 0000000000000000000000000000000000000000..d7fe57d898741d8ff0b7075fa5e307c2c2cf1833 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_pad_packed_sequence.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_pad_packed_sequence(Tensor data, Tensor batch_sizes, bool batch_first, Scalar padding_value, int total_length) -> (Tensor, Tensor) +inline ::std::tuple _pad_packed_sequence(const at::Tensor & data, const at::Tensor & batch_sizes, bool batch_first, const at::Scalar & padding_value, int64_t total_length) { + return at::_ops::_pad_packed_sequence::call(data, batch_sizes, batch_first, padding_value, total_length); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..dead21aa93e4f955b7f82ffd79c684f820be7670 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_scaled_dot_product_efficient_attention_backward_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple _scaled_dot_product_efficient_attention_backward_cuda(const at::Tensor & grad_out_, const at::Tensor & query, const at::Tensor & key, const at::Tensor & value, const at::Tensor & attn_bias, const at::Tensor & out, const at::Tensor & logsumexp, const at::Tensor & philox_seed, const at::Tensor & philox_offset, double dropout_p, ::std::array grad_input_mask, bool is_causal=false, c10::optional scale=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c3c728d69c4c30312f7dfcb00eb83399f41bfe4d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_softmax_backward_data_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API _softmax_backward_data { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, int64_t, at::ScalarType); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_softmax_backward_data") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_softmax_backward_data(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype) -> Tensor") + static at::Tensor call(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype); +}; + +struct TORCH_API _softmax_backward_data_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, int64_t, at::ScalarType, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::_softmax_backward_data") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "_softmax_backward_data.out(Tensor grad_output, Tensor output, int dim, ScalarType input_dtype, *, Tensor(a!) grad_input) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & grad_input); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad_output, const at::Tensor & output, int64_t dim, at::ScalarType input_dtype, at::Tensor & grad_input); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_native.h new file mode 100644 index 0000000000000000000000000000000000000000..66b85fa5ffb5c9ef5c890e0d6e2d0c8fa1bfd439 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_standard_gamma_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & _standard_gamma_out(const at::Tensor & self, c10::optional generator, at::Tensor & out); +TORCH_API at::Tensor _s_gamma_cpu(const at::Tensor & self, c10::optional generator=c10::nullopt); +TORCH_API at::Tensor _s_gamma_cuda(const at::Tensor & self, c10::optional generator=c10::nullopt); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsr.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsr.h new file mode 100644 index 0000000000000000000000000000000000000000..2753984047a9274a19a08494289e04520b73300b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_to_sparse_bsr.h @@ -0,0 +1,34 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::_to_sparse_bsr.out(Tensor self, int[2] blocksize, int? dense_dim=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _to_sparse_bsr_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim=c10::nullopt) { + return at::_ops::_to_sparse_bsr_out::call(self, blocksize, dense_dim, out); +} +// aten::_to_sparse_bsr.out(Tensor self, int[2] blocksize, int? dense_dim=None, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & _to_sparse_bsr_outf(const at::Tensor & self, at::IntArrayRef blocksize, c10::optional dense_dim, at::Tensor & out) { + return at::_ops::_to_sparse_bsr_out::call(self, blocksize, dense_dim, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_trilinear_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_trilinear_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..2ea991c7452b1511f08f0c2d44cac3c286050108 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_trilinear_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor _trilinear(const at::Tensor & i1, const at::Tensor & i2, const at::Tensor & i3, at::IntArrayRef expand1, at::IntArrayRef expand2, at::IntArrayRef expand3, at::IntArrayRef sumdim, int64_t unroll_dim=1); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..29dae2c5c84013fcf980c702a8d2b9d8bf96bae6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_unique2_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple _unique2(const at::Tensor & self, bool sorted=true, bool return_inverse=false, bool return_counts=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..d1ca316ac502dace3e7de9b322d2a6aea9ac8c4d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/_upsample_nearest_exact3d_backward_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor _upsample_nearest_exact3d_backward(const at::Tensor & grad_output, at::IntArrayRef output_size, at::IntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor _upsample_nearest_exact3d_backward_symint(const at::Tensor & grad_output, c10::SymIntArrayRef output_size, c10::SymIntArrayRef input_size, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..93b04891e650fa468be661ac36820ce3caa9f141 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_avg_pool2d_compositeimplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor adaptive_avg_pool2d(const at::Tensor & self, at::IntArrayRef output_size); +TORCH_API at::Tensor adaptive_avg_pool2d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_backward_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_backward_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..ca344b5ebe18466e734d683f09dda1cf55b87576 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/adaptive_max_pool2d_backward_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor adaptive_max_pool2d_backward(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices); +TORCH_API at::Tensor & adaptive_max_pool2d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices); +TORCH_API at::Tensor & adaptive_max_pool2d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, const at::Tensor & indices, at::Tensor & grad_input); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addmv_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addmv_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c1b3a1107a9e84f68c97ff2a4d95a5d5b7c1d609 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/addmv_meta_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor addmv(const at::Tensor & self, const at::Tensor & mat, const at::Tensor & vec, const at::Scalar & beta=1, const at::Scalar & alpha=1); +TORCH_API at::Tensor & addmv_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & mat, const at::Tensor & vec, const at::Scalar & beta=1, const at::Scalar & alpha=1); +TORCH_API at::Tensor & addmv_outf(const at::Tensor & self, const at::Tensor & mat, const at::Tensor & vec, const at::Scalar & beta, const at::Scalar & alpha, at::Tensor & out); +TORCH_API at::Tensor & addmv_(at::Tensor & self, const at::Tensor & mat, const at::Tensor & vec, const at::Scalar & beta=1, const at::Scalar & alpha=1); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/asin_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/asin_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..4c8dd2562d45a795ac586dba1b1a6ccf531bdda8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/asin_ops.h @@ -0,0 +1,50 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API asin { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::asin") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "asin(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API asin_ { + using schema = at::Tensor & (at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::asin_") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "asin_(Tensor(a!) self) -> Tensor(a!)") + static at::Tensor & call(at::Tensor & self); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, at::Tensor & self); +}; + +struct TORCH_API asin_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::asin") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "asin.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_depthwise3d_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_depthwise3d_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..6a647d1850f7008f6c9090aac95773998bd935de --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_depthwise3d_compositeexplicitautograd_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & conv_depthwise3d_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation); +TORCH_API at::Tensor & conv_depthwise3d_outf(const at::Tensor & self, const at::Tensor & weight, at::IntArrayRef kernel_size, const c10::optional & bias, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, at::Tensor & out); +TORCH_API at::Tensor & conv_depthwise3d_symint_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation); +TORCH_API at::Tensor & conv_depthwise3d_symint_outf(const at::Tensor & self, const at::Tensor & weight, c10::SymIntArrayRef kernel_size, const c10::optional & bias, c10::SymIntArrayRef stride, c10::SymIntArrayRef padding, c10::SymIntArrayRef dilation, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_tbc_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_tbc_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..31c307d4b307e5b17aa8065e3bb92bfc1be95a21 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/conv_tbc_backward_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::tuple conv_tbc_backward(const at::Tensor & self, const at::Tensor & input, const at::Tensor & weight, const at::Tensor & bias, int64_t pad); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copy_sparse_to_sparse_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copy_sparse_to_sparse_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0d265cdf8cebe5545d3be048da6dd65d9299046e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/copy_sparse_to_sparse_meta_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & copy_sparse_to_sparse_(at::Tensor & self, const at::Tensor & src, bool non_blocking=false); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/count_nonzero_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/count_nonzero_native.h new file mode 100644 index 0000000000000000000000000000000000000000..adeaae193e2eae5b2f7ced7cdf0a74f9eb57ef39 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/count_nonzero_native.h @@ -0,0 +1,25 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & count_nonzero_dim_IntList_out(const at::Tensor & self, at::IntArrayRef dim, at::Tensor & out); +TORCH_API at::Tensor count_nonzero_cpu(const at::Tensor & self, at::IntArrayRef dim); +TORCH_API at::Tensor count_nonzero_cuda(const at::Tensor & self, at::IntArrayRef dim); +TORCH_API at::Tensor count_nonzero(const at::Tensor & self, c10::optional dim=c10::nullopt); +TORCH_API at::Tensor & count_nonzero_out(const at::Tensor & self, c10::optional dim, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum.h new file mode 100644 index 0000000000000000000000000000000000000000..9844db25d7eba89e6c2d160bd5b297b2e8f14582 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum.h @@ -0,0 +1,53 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::cumsum(Tensor self, int dim, *, ScalarType? dtype=None) -> Tensor +inline at::Tensor cumsum(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt) { + return at::_ops::cumsum::call(self, dim, dtype); +} + +// aten::cumsum.out(Tensor self, int dim, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cumsum_out(at::Tensor & out, const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt) { + return at::_ops::cumsum_out::call(self, dim, dtype, out); +} +// aten::cumsum.out(Tensor self, int dim, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cumsum_outf(const at::Tensor & self, int64_t dim, c10::optional dtype, at::Tensor & out) { + return at::_ops::cumsum_out::call(self, dim, dtype, out); +} + +// aten::cumsum.dimname(Tensor self, Dimname dim, *, ScalarType? dtype=None) -> Tensor +inline at::Tensor cumsum(const at::Tensor & self, at::Dimname dim, c10::optional dtype=c10::nullopt) { + return at::_ops::cumsum_dimname::call(self, dim, dtype); +} + +// aten::cumsum.dimname_out(Tensor self, Dimname dim, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cumsum_out(at::Tensor & out, const at::Tensor & self, at::Dimname dim, c10::optional dtype=c10::nullopt) { + return at::_ops::cumsum_dimname_out::call(self, dim, dtype, out); +} +// aten::cumsum.dimname_out(Tensor self, Dimname dim, *, ScalarType? dtype=None, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & cumsum_outf(const at::Tensor & self, at::Dimname dim, c10::optional dtype, at::Tensor & out) { + return at::_ops::cumsum_dimname_out::call(self, dim, dtype, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..c237752891e49330b3518a55b0481093e3641c53 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/cumsum_cpu_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor cumsum(const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor & cumsum_out(at::Tensor & out, const at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); +TORCH_API at::Tensor & cumsum_outf(const at::Tensor & self, int64_t dim, c10::optional dtype, at::Tensor & out); +TORCH_API at::Tensor & cumsum_(at::Tensor & self, int64_t dim, c10::optional dtype=c10::nullopt); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..84324e29f311dcd5e218eb3d41d7536d644377dd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/detach_copy_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & detach_copy_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & detach_copy_outf(const at::Tensor & self, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..a93ff1ed3a43e0aecd6b9fff0aabd2e1f10328de --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/diagflat_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API diagflat { + using schema = at::Tensor (const at::Tensor &, int64_t); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::diagflat") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "diagflat(Tensor self, int offset=0) -> Tensor") + static at::Tensor call(const at::Tensor & self, int64_t offset); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t offset); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..822d606da502d5811e3f5d523045c6c72c475f6b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/digamma_cuda_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor digamma(const at::Tensor & self); +TORCH_API at::Tensor & digamma_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & digamma_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & digamma_(at::Tensor & self); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dist_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dist_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..37a9b79a90895bf526f334084a8dac55c668d008 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/dist_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API dist { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const at::Scalar &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::dist") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "dist(Tensor self, Tensor other, Scalar p=2) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other, const at::Scalar & p); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, const at::Scalar & p); +}; + +struct TORCH_API dist_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const at::Scalar &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::dist") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "dist.out(Tensor self, Tensor other, Scalar p=2, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, const at::Scalar & p, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, const at::Scalar & p, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/divide_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/divide_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..07e6a9c642aaaf5697ddb2dfee56291adee356ed --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/divide_compositeimplicitautograd_dispatch.h @@ -0,0 +1,34 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor divide(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & divide_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & divide_outf(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +TORCH_API at::Tensor & divide_(at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor divide(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & divide_(at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor divide(const at::Tensor & self, const at::Tensor & other, c10::optional rounding_mode); +TORCH_API at::Tensor & divide_out(at::Tensor & out, const at::Tensor & self, const at::Tensor & other, c10::optional rounding_mode); +TORCH_API at::Tensor & divide_outf(const at::Tensor & self, const at::Tensor & other, c10::optional rounding_mode, at::Tensor & out); +TORCH_API at::Tensor & divide_(at::Tensor & self, const at::Tensor & other, c10::optional rounding_mode); +TORCH_API at::Tensor divide(const at::Tensor & self, const at::Scalar & other, c10::optional rounding_mode); +TORCH_API at::Tensor & divide_(at::Tensor & self, const at::Scalar & other, c10::optional rounding_mode); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..f6b9aee974c9101d289672f4e7c1e75e052945e1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fake_quantize_per_channel_affine_cachemask_backward_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor fake_quantize_per_channel_affine_cachemask_backward(const at::Tensor & grad, const at::Tensor & mask); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_diagonal_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_diagonal_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..acaa70a8ced133e36b16de07480baee71130098d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fill_diagonal_compositeimplicitautograd_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor & fill_diagonal_(at::Tensor & self, const at::Scalar & fill_value, bool wrap=false); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fused_moving_avg_obs_fake_quant_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fused_moving_avg_obs_fake_quant_native.h new file mode 100644 index 0000000000000000000000000000000000000000..99bb467a249f1d4ad583592fa5ee25bf2682f9b4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/fused_moving_avg_obs_fake_quant_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor fused_moving_avg_obs_fake_quant(const at::Tensor & self, const at::Tensor & observer_on, const at::Tensor & fake_quant_on, at::Tensor & running_min, at::Tensor & running_max, at::Tensor & scale, at::Tensor & zero_point, double averaging_const, int64_t quant_min, int64_t quant_max, int64_t ch_axis, bool per_row_fake_quant=false, bool symmetric_quant=false); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_backward_jvp.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_backward_jvp.h new file mode 100644 index 0000000000000000000000000000000000000000..bdc71827cac964506d9cfd79a056f74332ed1f89 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/glu_backward_jvp.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::glu_backward_jvp(Tensor grad_x, Tensor grad_glu, Tensor x, Tensor dgrad_glu, Tensor dx, int dim) -> Tensor +inline at::Tensor glu_backward_jvp(const at::Tensor & grad_x, const at::Tensor & grad_glu, const at::Tensor & x, const at::Tensor & dgrad_glu, const at::Tensor & dx, int64_t dim) { + return at::_ops::glu_backward_jvp::call(grad_x, grad_glu, x, dgrad_glu, dx, dim); +} + +// aten::glu_backward_jvp.out(Tensor grad_x, Tensor grad_glu, Tensor x, Tensor dgrad_glu, Tensor dx, int dim, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & glu_backward_jvp_out(at::Tensor & out, const at::Tensor & grad_x, const at::Tensor & grad_glu, const at::Tensor & x, const at::Tensor & dgrad_glu, const at::Tensor & dx, int64_t dim) { + return at::_ops::glu_backward_jvp_out::call(grad_x, grad_glu, x, dgrad_glu, dx, dim, out); +} +// aten::glu_backward_jvp.out(Tensor grad_x, Tensor grad_glu, Tensor x, Tensor dgrad_glu, Tensor dx, int dim, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & glu_backward_jvp_outf(const at::Tensor & grad_x, const at::Tensor & grad_glu, const at::Tensor & x, const at::Tensor & dgrad_glu, const at::Tensor & dx, int64_t dim, at::Tensor & out) { + return at::_ops::glu_backward_jvp_out::call(grad_x, grad_glu, x, dgrad_glu, dx, dim, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_backward_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_backward_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..540de004e7a63a7d304da9636d58c94701a3d9cf --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/grid_sampler_2d_backward_cuda_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple grid_sampler_2d_backward(const at::Tensor & grad_output, const at::Tensor & input, const at::Tensor & grid, int64_t interpolation_mode, int64_t padding_mode, bool align_corners, ::std::array output_mask); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_native.h new file mode 100644 index 0000000000000000000000000000000000000000..b61d0dabb303f36984bf3d908be456045bbc58e8 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/gt_native.h @@ -0,0 +1,31 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace native { +struct TORCH_API structured_gt_Scalar_out : public at::meta::structured_gt_Scalar { +void impl(const at::Tensor & self, const at::Scalar & other, const at::Tensor & out); +}; +TORCH_API at::Tensor gt_scalar_nested(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor gt_quantized_cpu(const at::Tensor & self, const at::Scalar & other); +TORCH_API at::Tensor & gt_out_quantized_cpu(const at::Tensor & self, const at::Scalar & other, at::Tensor & out); +struct TORCH_API structured_gt_Tensor_out : public at::meta::structured_gt_Tensor { +void impl(const at::Tensor & self, const at::Tensor & other, const at::Tensor & out); +}; +TORCH_API at::Tensor gt_quantized_cpu(const at::Tensor & self, const at::Tensor & other); +TORCH_API at::Tensor & gt_out_quantized_cpu(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8dfe7a032a380e0ef452f5a30f6a187b07135744 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardshrink_backward_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor hardshrink_backward(const at::Tensor & grad_out, const at::Tensor & self, const at::Scalar & lambd); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh.h new file mode 100644 index 0000000000000000000000000000000000000000..764c572659fac897a22fc6622b873ea006f9d14d --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/hardtanh.h @@ -0,0 +1,44 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::hardtanh.out(Tensor self, Scalar min_val=-1, Scalar max_val=1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & hardtanh_out(at::Tensor & out, const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1) { + return at::_ops::hardtanh_out::call(self, min_val, max_val, out); +} +// aten::hardtanh.out(Tensor self, Scalar min_val=-1, Scalar max_val=1, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & hardtanh_outf(const at::Tensor & self, const at::Scalar & min_val, const at::Scalar & max_val, at::Tensor & out) { + return at::_ops::hardtanh_out::call(self, min_val, max_val, out); +} + +// aten::hardtanh(Tensor self, Scalar min_val=-1, Scalar max_val=1) -> Tensor +inline at::Tensor hardtanh(const at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1) { + return at::_ops::hardtanh::call(self, min_val, max_val); +} + +// aten::hardtanh_(Tensor(a!) self, Scalar min_val=-1, Scalar max_val=1) -> Tensor(a!) +inline at::Tensor & hardtanh_(at::Tensor & self, const at::Scalar & min_val=-1, const at::Scalar & max_val=1) { + return at::_ops::hardtanh_::call(self, min_val, max_val); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..724242bd5929ba52fe380f361cbdbe6cb32b9a6a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/indices_copy_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API indices_copy { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::indices_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "indices_copy(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API indices_copy_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::indices_copy") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "indices_copy.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_signed.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_signed.h new file mode 100644 index 0000000000000000000000000000000000000000..decbb4e5a1622220c9517ac286d85f7426581d76 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/is_signed.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::is_signed(Tensor self) -> bool +inline bool __dispatch_is_signed(const at::Tensor & self) { + return at::_ops::is_signed::call(self); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isposinf_meta.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isposinf_meta.h new file mode 100644 index 0000000000000000000000000000000000000000..c3913996da1e354b484559d7d12eb88ebdabdf88 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/isposinf_meta.h @@ -0,0 +1,27 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeMetaFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +namespace at { +namespace meta { + +struct TORCH_API structured_isposinf : public TensorIteratorBase { + + + void meta(const at::Tensor & self); +}; + +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eig_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eig_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f5b1d64d56ef10a989a2b7ebfb99dfedb104ceb3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linalg_eig_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API ::std::tuple linalg_eig(const at::Tensor & self); +TORCH_API ::std::tuple linalg_eig_out(at::Tensor & eigenvalues, at::Tensor & eigenvectors, const at::Tensor & self); +TORCH_API ::std::tuple linalg_eig_outf(const at::Tensor & self, at::Tensor & eigenvalues, at::Tensor & eigenvectors); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..8524ed838282448ce502ace50c14d85cc4ec45b2 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_backward_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API linear_backward { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, const at::Tensor &, ::std::array); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linear_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linear_backward(Tensor self, Tensor grad_output, Tensor weight, bool[3] output_mask) -> (Tensor, Tensor, Tensor)") + static ::std::tuple call(const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, ::std::array output_mask); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, ::std::array output_mask); +}; + +struct TORCH_API linear_backward_out { + using schema = ::std::tuple (const at::Tensor &, const at::Tensor &, const at::Tensor &, ::std::array, at::Tensor &, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::linear_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "linear_backward.out(Tensor self, Tensor grad_output, Tensor weight, bool[3] output_mask, *, Tensor(a!) out0, Tensor(b!) out1, Tensor(c!) out2) -> (Tensor(a!), Tensor(b!), Tensor(c!))") + static ::std::tuple call(const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & grad_output, const at::Tensor & weight, ::std::array output_mask, at::Tensor & out0, at::Tensor & out1, at::Tensor & out2); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_native.h new file mode 100644 index 0000000000000000000000000000000000000000..789ce446ff1e1dc3381fc18e402f9f950a792a08 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/linear_native.h @@ -0,0 +1,23 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor linear(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias={}); +TORCH_API at::Tensor & linear_out(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias, at::Tensor & out); +TORCH_API at::Tensor nested_linear(const at::Tensor & input, const at::Tensor & weight, const c10::optional & bias={}); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log2_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log2_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..0c3163a4ff9fd5eb0c470e199be0f7aec2fb4f50 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/log2_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor log2(const at::Tensor & self); +TORCH_API at::Tensor & log2_(at::Tensor & self); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_fill_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_fill_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..eb04f83190a57ebe8eb34d584f3648dc2c5830d1 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_fill_meta_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor & masked_fill_(at::Tensor & self, const at::Tensor & mask, const at::Scalar & value); +TORCH_API at::Tensor & masked_fill_(at::Tensor & self, const at::Tensor & mask, const at::Tensor & value); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_backward_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_backward_native.h new file mode 100644 index 0000000000000000000000000000000000000000..cafa2d535da6476e9273b0f976bab849689ed425 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/masked_scatter_backward_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor masked_scatter_backward_symint(const at::Tensor & grad_output, const at::Tensor & mask, c10::SymIntArrayRef sizes); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..92d8ae8c0a3ebd4c91a81226dad5f7b39053919a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_ops.h @@ -0,0 +1,105 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API max_dim { + using schema = ::std::tuple (const at::Tensor &, int64_t, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dim") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.dim(Tensor self, int dim, bool keepdim=False) -> (Tensor values, Tensor indices)") + static ::std::tuple call(const at::Tensor & self, int64_t dim, bool keepdim); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool keepdim); +}; + +struct TORCH_API max_dim_max { + using schema = ::std::tuple (const at::Tensor &, int64_t, bool, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "dim_max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.dim_max(Tensor self, int dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)") + static ::std::tuple call(const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & max, at::Tensor & max_values); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, int64_t dim, bool keepdim, at::Tensor & max, at::Tensor & max_values); +}; + +struct TORCH_API max_names_dim { + using schema = ::std::tuple (const at::Tensor &, at::Dimname, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names_dim") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.names_dim(Tensor self, Dimname dim, bool keepdim=False) -> (Tensor values, Tensor indices)") + static ::std::tuple call(const at::Tensor & self, at::Dimname dim, bool keepdim); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, bool keepdim); +}; + +struct TORCH_API max_names_dim_max { + using schema = ::std::tuple (const at::Tensor &, at::Dimname, bool, at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "names_dim_max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.names_dim_max(Tensor self, Dimname dim, bool keepdim=False, *, Tensor(a!) max, Tensor(b!) max_values) -> (Tensor(a!) values, Tensor(b!) indices)") + static ::std::tuple call(const at::Tensor & self, at::Dimname dim, bool keepdim, at::Tensor & max, at::Tensor & max_values); + static ::std::tuple redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Dimname dim, bool keepdim, at::Tensor & max, at::Tensor & max_values); +}; + +struct TORCH_API max { + using schema = at::Tensor (const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max(Tensor self) -> Tensor") + static at::Tensor call(const at::Tensor & self); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self); +}; + +struct TORCH_API max_other { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "other") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.other(Tensor self, Tensor other) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & other); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other); +}; + +struct TORCH_API max_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.out(Tensor self, Tensor other, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & other, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & other, at::Tensor & out); +}; + +struct TORCH_API max_unary_out { + using schema = at::Tensor & (const at::Tensor &, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::max") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "unary_out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "max.unary_out(Tensor self, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_backward_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_backward_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..3ff969183a4b7be6dac728dbc965f73e89b12012 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/max_pool2d_with_indices_backward_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor max_pool2d_with_indices_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, const at::Tensor & indices); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_depthwise_convolution_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_depthwise_convolution_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..f6756336099b540fda563bf8111d0496c778309c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/miopen_depthwise_convolution_ops.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API miopen_depthwise_convolution { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, const c10::optional &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymInt, bool, bool); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::miopen_depthwise_convolution") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "miopen_depthwise_convolution(Tensor self, Tensor weight, Tensor? bias, SymInt[] padding, SymInt[] stride, SymInt[] dilation, SymInt groups, bool benchmark, bool deterministic) -> Tensor") + static at::Tensor call(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic); +}; + +struct TORCH_API miopen_depthwise_convolution_out { + using schema = at::Tensor & (const at::Tensor &, const at::Tensor &, const c10::optional &, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymIntArrayRef, c10::SymInt, bool, bool, at::Tensor &); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::miopen_depthwise_convolution") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "out") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "miopen_depthwise_convolution.out(Tensor self, Tensor weight, Tensor? bias, SymInt[] padding, SymInt[] stride, SymInt[] dilation, SymInt groups, bool benchmark, bool deterministic, *, Tensor(a!) out) -> Tensor(a!)") + static at::Tensor & call(const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, at::Tensor & out); + static at::Tensor & redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & self, const at::Tensor & weight, const c10::optional & bias, c10::SymIntArrayRef padding, c10::SymIntArrayRef stride, c10::SymIntArrayRef dilation, c10::SymInt groups, bool benchmark, bool deterministic, at::Tensor & out); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_native.h new file mode 100644 index 0000000000000000000000000000000000000000..d47bb8a2978f8b2f435852fb98dbcba65bfe1a16 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantize_per_tensor_native.h @@ -0,0 +1,26 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor & quantize_per_tensor_out(const at::Tensor & self, double scale, int64_t zero_point, at::ScalarType dtype, at::Tensor & out); +TORCH_API at::Tensor quantize_per_tensor(const at::Tensor & self, double scale, int64_t zero_point, at::ScalarType dtype); +TORCH_API at::Tensor & quantize_per_tensor_tensor_qparams_out(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, at::ScalarType dtype, at::Tensor & out); +TORCH_API at::Tensor quantize_per_tensor_tensor_qparams(const at::Tensor & self, const at::Tensor & scale, const at::Tensor & zero_point, at::ScalarType dtype); +TORCH_API void quantize_per_tensor_tensors_out(at::TensorList tensors, const at::Tensor & scales, const at::Tensor & zero_points, at::ScalarType dtype, at::TensorList out); +TORCH_API ::std::vector quantize_per_tensor_list_cpu(at::TensorList tensors, const at::Tensor & scales, const at::Tensor & zero_points, at::ScalarType dtype); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool3d_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool3d_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..fe63b83559d48d5140ef44a2604a4cc5c05a8613 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/quantized_max_pool3d_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor & quantized_max_pool3d_out(at::Tensor & out, const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride={}, at::IntArrayRef padding=0, at::IntArrayRef dilation=1, bool ceil_mode=false); +TORCH_API at::Tensor & quantized_max_pool3d_outf(const at::Tensor & self, at::IntArrayRef kernel_size, at::IntArrayRef stride, at::IntArrayRef padding, at::IntArrayRef dilation, bool ceil_mode, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reciprocal_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reciprocal_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e5df2fd6424e49f31a74e1c18a915610620015aa --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reciprocal_cpu_dispatch.h @@ -0,0 +1,26 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor reciprocal(const at::Tensor & self); +TORCH_API at::Tensor & reciprocal_out(at::Tensor & out, const at::Tensor & self); +TORCH_API at::Tensor & reciprocal_outf(const at::Tensor & self, at::Tensor & out); +TORCH_API at::Tensor & reciprocal_(at::Tensor & self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward.h new file mode 100644 index 0000000000000000000000000000000000000000..4e8ffdf9ad4b4043be13a3497e36fc6519460c0b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward.h @@ -0,0 +1,91 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::reflection_pad1d_backward.grad_input(Tensor grad_output, Tensor self, SymInt[2] padding, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & reflection_pad1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, c10::fromIntArrayRefSlow(padding), grad_input); +} +namespace symint { + template ::value>> + at::Tensor & reflection_pad1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, c10::fromIntArrayRefSlow(padding), grad_input); + } +} + +// aten::reflection_pad1d_backward.grad_input(Tensor grad_output, Tensor self, SymInt[2] padding, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & reflection_pad1d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding, at::Tensor & grad_input) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, c10::fromIntArrayRefSlow(padding), grad_input); +} +namespace symint { + template ::value>> + at::Tensor & reflection_pad1d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding, at::Tensor & grad_input) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, c10::fromIntArrayRefSlow(padding), grad_input); + } +} + +// aten::reflection_pad1d_backward.grad_input(Tensor grad_output, Tensor self, SymInt[2] padding, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & reflection_pad1d_backward_symint_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, padding, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & reflection_pad1d_backward_out(at::Tensor & grad_input, const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, padding, grad_input); + } +} + +// aten::reflection_pad1d_backward.grad_input(Tensor grad_output, Tensor self, SymInt[2] padding, *, Tensor(a!) grad_input) -> Tensor(a!) +inline at::Tensor & reflection_pad1d_backward_symint_outf(const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & grad_input) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, padding, grad_input); +} +namespace symint { + template ::value>> + at::Tensor & reflection_pad1d_backward_outf(const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding, at::Tensor & grad_input) { + return at::_ops::reflection_pad1d_backward_grad_input::call(grad_output, self, padding, grad_input); + } +} + +// aten::reflection_pad1d_backward(Tensor grad_output, Tensor self, SymInt[2] padding) -> Tensor +inline at::Tensor reflection_pad1d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding) { + return at::_ops::reflection_pad1d_backward::call(grad_output, self, c10::fromIntArrayRefSlow(padding)); +} +namespace symint { + template ::value>> + at::Tensor reflection_pad1d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding) { + return at::_ops::reflection_pad1d_backward::call(grad_output, self, c10::fromIntArrayRefSlow(padding)); + } +} + +// aten::reflection_pad1d_backward(Tensor grad_output, Tensor self, SymInt[2] padding) -> Tensor +inline at::Tensor reflection_pad1d_backward_symint(const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding) { + return at::_ops::reflection_pad1d_backward::call(grad_output, self, padding); +} +namespace symint { + template ::value>> + at::Tensor reflection_pad1d_backward(const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding) { + return at::_ops::reflection_pad1d_backward::call(grad_output, self, padding); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..497a7af3b3b2151f5145b751b79528f8e7d8192b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/reflection_pad1d_backward_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor reflection_pad1d_backward(const at::Tensor & grad_output, const at::Tensor & self, at::IntArrayRef padding); +TORCH_API at::Tensor reflection_pad1d_backward_symint(const at::Tensor & grad_output, const at::Tensor & self, c10::SymIntArrayRef padding); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..8559af53004d4c056c4ffb3d92a37b4eaa85c831 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/select_compositeexplicitautograd_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor select(const at::Tensor & self, int64_t dim, int64_t index); +TORCH_API at::Tensor select_symint(const at::Tensor & self, int64_t dim, c10::SymInt index); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu.h new file mode 100644 index 0000000000000000000000000000000000000000..6aa8859693a9f6f9ac9dd89754ee19abffddae6a --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/silu.h @@ -0,0 +1,44 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::silu(Tensor self) -> Tensor +inline at::Tensor silu(const at::Tensor & self) { + return at::_ops::silu::call(self); +} + +// aten::silu_(Tensor(a!) self) -> Tensor(a!) +inline at::Tensor & silu_(at::Tensor & self) { + return at::_ops::silu_::call(self); +} + +// aten::silu.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & silu_out(at::Tensor & out, const at::Tensor & self) { + return at::_ops::silu_out::call(self, out); +} +// aten::silu.out(Tensor self, *, Tensor(a!) out) -> Tensor(a!) +inline at::Tensor & silu_outf(const at::Tensor & self, at::Tensor & out) { + return at::_ops::silu_out::call(self, out); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_inverse_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_inverse_native.h new file mode 100644 index 0000000000000000000000000000000000000000..9ef58c72d95a8d2fdb44399ddf974947d43e2b5b --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/slice_inverse_native.h @@ -0,0 +1,21 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API at::Tensor slice_inverse_symint(const at::Tensor & self, const at::Tensor & src, int64_t dim=0, c10::optional start=c10::nullopt, c10::optional end=c10::nullopt, c10::SymInt step=1); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_chebyshev_polynomial_u_compositeexplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_chebyshev_polynomial_u_compositeexplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..e8e0a0d8ae0d1925f6e874fd156fa5a5e941ab0c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_chebyshev_polynomial_u_compositeexplicitautograd_dispatch.h @@ -0,0 +1,28 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautograd { + +TORCH_API at::Tensor special_chebyshev_polynomial_u(const at::Scalar & x, const at::Tensor & n); +TORCH_API at::Tensor & special_chebyshev_polynomial_u_out(at::Tensor & out, const at::Scalar & x, const at::Tensor & n); +TORCH_API at::Tensor & special_chebyshev_polynomial_u_outf(const at::Scalar & x, const at::Tensor & n, at::Tensor & out); +TORCH_API at::Tensor special_chebyshev_polynomial_u(const at::Tensor & x, const at::Scalar & n); +TORCH_API at::Tensor & special_chebyshev_polynomial_u_out(at::Tensor & out, const at::Tensor & x, const at::Scalar & n); +TORCH_API at::Tensor & special_chebyshev_polynomial_u_outf(const at::Tensor & x, const at::Scalar & n, at::Tensor & out); + +} // namespace compositeexplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_meta_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_meta_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..4b0ae95104ca4b441bc8ef7b57f3119958b64b7c --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_hermite_polynomial_he_meta_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace meta { + +TORCH_API at::Tensor special_hermite_polynomial_he(const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_hermite_polynomial_he_out(at::Tensor & out, const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_hermite_polynomial_he_outf(const at::Tensor & x, const at::Tensor & n, at::Tensor & out); + +} // namespace meta +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_compositeimplicitautograd_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_compositeimplicitautograd_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..932694c6006acc11d4147ba9f864d6588e44b887 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_round_compositeimplicitautograd_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeimplicitautograd { + +TORCH_API at::Tensor special_round(const at::Tensor & self, int64_t decimals=0); +TORCH_API at::Tensor & special_round_out(at::Tensor & out, const at::Tensor & self, int64_t decimals=0); +TORCH_API at::Tensor & special_round_outf(const at::Tensor & self, int64_t decimals, at::Tensor & out); + +} // namespace compositeimplicitautograd +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_scaled_modified_bessel_k0_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_scaled_modified_bessel_k0_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..f2b3f8ddf071b56969ecfa467d4a162e5ca3ceb6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_scaled_modified_bessel_k0_cpu_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor special_scaled_modified_bessel_k0(const at::Tensor & x); +TORCH_API at::Tensor & special_scaled_modified_bessel_k0_out(at::Tensor & out, const at::Tensor & x); +TORCH_API at::Tensor & special_scaled_modified_bessel_k0_outf(const at::Tensor & x, at::Tensor & out); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_t_cuda_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_t_cuda_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..38796f7b69eaafa55709ac69ff24694d1222b4fd --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/special_shifted_chebyshev_polynomial_t_cuda_dispatch.h @@ -0,0 +1,25 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cuda { + +TORCH_API at::Tensor special_shifted_chebyshev_polynomial_t(const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_t_out(at::Tensor & out, const at::Tensor & x, const at::Tensor & n); +TORCH_API at::Tensor & special_shifted_chebyshev_polynomial_t_outf(const at::Tensor & x, const at::Tensor & n, at::Tensor & out); + +} // namespace cuda +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size.h new file mode 100644 index 0000000000000000000000000000000000000000..e7cfa73c4bbeebc268ad532e625b9297f66d1478 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/sum_to_size.h @@ -0,0 +1,39 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +namespace symint { + template ::value>> + at::Tensor sum_to_size(const at::Tensor & self, at::IntArrayRef size) { + return at::_ops::sum_to_size::call(self, c10::fromIntArrayRefSlow(size)); + } +} + +namespace symint { + template ::value>> + at::Tensor sum_to_size(const at::Tensor & self, c10::SymIntArrayRef size) { + return at::_ops::sum_to_size::call(self, size); + } +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/swapaxes.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/swapaxes.h new file mode 100644 index 0000000000000000000000000000000000000000..57a63505218ad7d017e8a770f27bac84bcb32c12 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/swapaxes.h @@ -0,0 +1,30 @@ +#pragma once + +// @generated by torchgen/gen.py from Function.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + + +#include + +namespace at { + + +// aten::swapaxes(Tensor(a) self, int axis0, int axis1) -> Tensor(a) +inline at::Tensor swapaxes(const at::Tensor & self, int64_t axis0, int64_t axis1) { + return at::_ops::swapaxes::call(self, axis0, axis1); +} + +} diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_dense_backward_ops.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_dense_backward_ops.h new file mode 100644 index 0000000000000000000000000000000000000000..c219291a8b858306cf4c5a128d0c3d685a2e47f4 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/to_dense_backward_ops.h @@ -0,0 +1,28 @@ +#pragma once + +// @generated by torchgen/gen.py from Operator.h + +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { +namespace _ops { + + +struct TORCH_API to_dense_backward { + using schema = at::Tensor (const at::Tensor &, const at::Tensor &, c10::optional); + using ptr_schema = schema*; + // See Note [static constexpr char* members for windows NVCC] + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(name, "aten::to_dense_backward") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(overload_name, "") + STATIC_CONSTEXPR_STR_INL_EXCEPT_WIN_CUDA(schema_str, "to_dense_backward(Tensor grad, Tensor input, bool? masked_grad=None) -> Tensor") + static at::Tensor call(const at::Tensor & grad, const at::Tensor & input, c10::optional masked_grad); + static at::Tensor redispatch(c10::DispatchKeySet dispatchKeySet, const at::Tensor & grad, const at::Tensor & input, c10::optional masked_grad); +}; + +}} // namespace at::_ops diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..1d87e5101da56c8087d383f7ba76981d8c2284f6 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/trace_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API at::Tensor trace(const at::Tensor & self); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_cpu_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_cpu_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..100f6b30a01dd648ea02d67d20b3ffd20a7e25e7 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unique_dim_consecutive_cpu_dispatch.h @@ -0,0 +1,23 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace cpu { + +TORCH_API ::std::tuple unique_dim_consecutive(const at::Tensor & self, int64_t dim, bool return_inverse=false, bool return_counts=false); + +} // namespace cpu +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unsafe_split_with_sizes_native.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unsafe_split_with_sizes_native.h new file mode 100644 index 0000000000000000000000000000000000000000..34750e95124062b9c10ab58569b8d962d79354c3 --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/unsafe_split_with_sizes_native.h @@ -0,0 +1,22 @@ +#pragma once + +// @generated by torchgen/gen.py from NativeFunction.h + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + + +namespace at { +namespace native { +TORCH_API ::std::vector unsafe_split_with_sizes(const at::Tensor & self, at::IntArrayRef split_sizes, int64_t dim=0); +TORCH_API void unsafe_split_with_sizes_out_symint(const at::Tensor & self, c10::SymIntArrayRef split_sizes, int64_t dim, at::TensorList out); +} // namespace native +} // namespace at diff --git a/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_compositeexplicitautogradnonfunctional_dispatch.h b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_compositeexplicitautogradnonfunctional_dispatch.h new file mode 100644 index 0000000000000000000000000000000000000000..dd794485a332567de08fd6c66423ea36b88f732e --- /dev/null +++ b/tuning-competition-baseline/.venv/lib/python3.11/site-packages/torch/include/ATen/ops/upsample_trilinear3d_compositeexplicitautogradnonfunctional_dispatch.h @@ -0,0 +1,24 @@ +#pragma once +// @generated by torchgen/gen.py from DispatchKeyFunction.h + +// NB: The implementing C++ file is RegisterDispatchKey.cpp + +// The only #includes we need are for custom classes that have defaults in the C++ API +#include +#include +#include + +// Forward declarations of any types needed in the operator signatures. +// We can't directly include these classes because it will cause circular include dependencies. +// This file is included by TensorBody.h, which defines the Tensor class. +#include + +namespace at { + +namespace compositeexplicitautogradnonfunctional { + +TORCH_API at::Tensor upsample_trilinear3d(const at::Tensor & self, at::IntArrayRef output_size, bool align_corners, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); +TORCH_API at::Tensor upsample_trilinear3d_symint(const at::Tensor & self, c10::SymIntArrayRef output_size, bool align_corners, c10::optional scales_d=c10::nullopt, c10::optional scales_h=c10::nullopt, c10::optional scales_w=c10::nullopt); + +} // namespace compositeexplicitautogradnonfunctional +} // namespace at